- Email: [email protected]
Smoking increases the risk of delirium for older inpatients: a prospective population-based study
General Hospital Psychiatry 37 (2015) 360–364
Contents lists available at ScienceDirect
General Hospital Psychiatry journal homepage: http://www.ghpjournal.com
Smoking increases the risk of delirium for older inpatients: a prospective population-based study Johannes Baltasar Hessler, M.Sc. a,⁎, Monika Brönner, M.D. a, Thorleif Etgen, M.D. a,b, Othmar Gotzler, M.D. c, Hans Förstl, M.D. a, Holger Poppert, M.D. d, Dirk Sander, M.D. e, Horst Bickel, Ph.D. a a
Department of Psychiatry and Psychotherapy, Technische Universität München Klinikum Rechts der Isar, Ismaninger Strasse 22, 81675 Munich, Germany Department of Neurology, Kliniken Südostbayern–Klinikum Traunstein, Cuno-Niggl-Strasse 3, 83278 Traunstein, Germany INVADE Study Group, Karl-Böhm-Strasse 32, 85598 Baldham, Germany d Department of Neurology, Technische Universität München Klinikum Rechts der Isar, Ismaninger Strasse 22, 81675 Munich, Germany e Department of Neurology, Benedictus Krankenhaus Tutzing, Bahnhofstrasse 5, 82327 Tutzing, Germany b c
a r t i c l e
i n f o
Article history: Received 26 November 2014 Revised 13 March 2015 Accepted 13 March 2015 Keywords: Acetylcholine Nicotine Withdrawal Prevention Hospital
a b s t r a c t Objectives: To investigate the association between smoking in the older population and the risk of inpatient delirium, which is common and has adverse consequences. Method: Participants (N=3754) were insurants aged ≥55 years of the largest German statutory health insurance company, who enrolled in a 6-year prospective population-based study. Baseline smoking, adjusted for age, sex, depressive symptoms, cognitive impairment and alcohol consumption, was analyzed as risk factor of inpatient delirium. Results are presented as hazard ratios (HRs) and 95% confidence intervals (95% CIs). Results: Three-hundred seventy-three (10.0%) participants were smokers at baseline, 865 (23.0%) were quitters and 2516 (67.0%) were lifelong abstainers. Mean pack-years of smokers and quitters were 23.8 (S.D.=22.4). Sixty-one (1.6%) received a diagnosis of inpatient delirium. Smokers had an increased risk of delirium compared to abstainers in the fully adjusted model (HR=2.87, 95% CI 1.24–6.66). Quitters and abstainers did not differ (HR=0.79, 95% CI 0.37–1.72). Comparing smokers and quitters, current smoking status (HR=3.22, 95% CI 1.20–8.62) but not pack-years [residual χ2(1)=0.25, P=.874] were associated with inpatient delirium. Conclusion: Only current smoking but not being a quitter and the lifetime amount smoked were associated with inpatient delirium, indicating that acute nicotine withdrawal may represent a relevant pathogenic mechanism. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Delirium describes acute brain failure that causes impairment of attention and other cognitive functions [1]. The occurrence of delirium appears to result from the interplay of a wide range of precipitating noxious events and predisposing factors [2]. Among others, high age; cognitive, functional and sensory impairment; infection; physiological abnormalities; a history of physical and mental disease; drug use; alcohol misuse; physical constraints; anesthesia; use of a bladder catheter; urgent or trauma admission and intensive care unit (ICU) stay can contribute to the etiology of delirium [3,4]. Delirium is most common among older hospital patients [5], with the highest incidence rates in surgical and ICU patients [3]. Given the high costs, prolonged hospital stays and increased rates of morbidity and mortality associated with ⁎ Corresponding author. Department of Psychiatry and Psychotherapy, Klinikum Rechts der Isar der Technische Universität München, Ismaninger Strasse 22, 81675 Munich, Germany. Tel.: +49-89-4140-6183; fax: +49-89-4140-6379. E-mail addresses: [email protected] (J.B. Hessler), [email protected] (M. Brönner), [email protected] (T. Etgen), [email protected] (O. Gotzler), [email protected] (H. Förstl), [email protected] (H. Poppert), [email protected] (D. Sander), [email protected] (H. Bickel). http://dx.doi.org/10.1016/j.genhosppsych.2015.03.009 0163-8343/© 2015 Elsevier Inc. All rights reserved.
delirium [1], as well as the poor cognitive and functional prognosis for those affected [6], the investigation of means to reduce its incidence in older hospital patients is of vital importance. Several multicomponent prevention strategies have been proven effective in reducing the incidence of delirium in hospitalized older people [7,8]. Surprisingly, since being a common habit, smoking has not been considered in this context. This neglect is possibly due to the sparse and inconsistent evidence on the role of smoking in the pathogenesis of delirium [9]. In some studies, current smoking was found to increase the risk of delirium in the hospital [10–13], yet others found no effect [14–17]. Despite the lack of clear evidence, there is a reason to believe that smoking could be a relevant factor in the development of delirium, both as predisposing factor and precipitating noxious event. Next to other factors (e.g., hypoxemia, increased carbon monoxide levels), smoking potentially increases the risk of delirium through the buildup of microvascular and atherosclerotic changes in the brain [11,18,19]. These changes are positively correlated with the number of cigarettes smoked and can persist for many years after quitting [20]. If vascular changes were the main agent behind the association between smoking and delirium, not current smoking status but the lifetime number of cigarettes smoked should predict the occurrence of delirium. Case
J.B. Hessler et al. / General Hospital Psychiatry 37 (2015) 360–364
reports of heavily smoking hospital patients [21–24], however, point to an alternative or additional pathway from smoking to delirium. These patients had to stop smoking due to their treatment and developed delirium. The symptoms remitted within hours after the application of transdermal nicotine replacement. These reports suggest that acute nicotine withdrawal due to smoking cessation in the hospital may constitute a noxious event, which directly relates to the development of delirium. If nicotine withdrawal and not atherosclerosis were the critical factor linking smoking and delirium, current smoking status but not the number of cigarettes smoked should predict incident delirium. The present study aimed to investigate the unclear association between smoking in the older population and the incidence of inpatient delirium, as well as explore possible underlying mechanisms. To this purpose, we analyzed data of almost 4000 individuals older than 54 years who were prospectively observed for an average of 6 years over the course of the INVADE trial (Intervention Project on Cerebrovascular Disease and Dementia in the District of Ebersberg [25]). In particular, we examined two questions. (1) Do quitters and current smokers differ from abstainers in their risk of inpatient delirium? (2) Is smoking status or the lifetime amount smoked the better predictor of inpatient delirium among current smokers and quitters? 2. Method 2.1. Participants The INVADE trial is a prospective and population-based cohort study in a geographically defined area. Participants were identified from the database of the statutory health insurance company Allgemeine Ortskrankenkasse (AOK). Membership in a health insurance is mandatory in Germany and AOK holds the largest market share, representing around 40% of the total population. In 2001, 11,317 insurants met the inclusion criteria of being 55 years or older, as well as living in the district of Ebersberg, and were invited to participate. Between 2001 and 2003, 3908 participants enrolled in the trial. 2.2. Procedure The ethics committee of the Faculty of Medicine at the Technische Universität München approved the study protocol and all participants signed informed consent. The participants were examined by their respective general practitioner (GP) at baseline and at follow-ups after 2, 4 and 6 years. At baseline, GPs asked the participants about their current and previous smoking habits. Based on this information, the participants were categorized into current smokers, quitters and lifelong abstainers. For smokers and quitters, the number of pack-years was determined as an indicator of the lifetime amount smoked. Pack-years are calculated by multiplying the number of packs a person smoked per day with the number of years the person has smoked. Further, the participants filled in questionnaires about sociodemographic data, subjective health, use of medical services, memory complaints and depressive symptoms (Geriatric Depression Scale or GDS [26]). GPs reported the patients’ previous and current diagnoses, current medication, alcohol consumption, physical activity, body mass index (BMI), impairment of activities of daily living (Rankin Scale [27]), blood pressure, ankle-to-brachial index (ABI), and cognitive status (6-Item Cognitive Impairment Test or 6CIT [28,29]) and conducted an electrocardiogram. GPs also took fasting blood samples that were analyzed in a central laboratory with regard to total cholesterol, low- and high-density lipoprotein (LDL, HDL) cholesterol, triglycerides, serum glucose, glycosylated hemoglobin A1c (HbA1c), creatinine, homocysteine and high-sensitivity C-reactive protein (hsCRP). A detailed description of the INVADE trial and the baseline examination is published [25]. Cases of delirium not induced by alcohol and other psychoactive substances were identified by searching claims data of the AOK health
361
insurance for International Statistical Classification of Diseases (ICD), 10th Revision codes F05.0–F05.9 that were assigned during a hospital stay over the course of the 6-year period of the INVADE trial. If participants received more than one diagnosis of delirium over the course of the study, the first diagnosis was used as endpoint. Diagnoses of substance withdrawal state with delirium (F1×.4) were not considered.
2.3. Statistical analysis In order to establish an independent association between smoking and inpatient delirium, the known risk factors age, sex, cognitive impairment (6CIT), depressive symptoms (GDS), alcohol consumption (GP), use of psychoactive drugs (GP) and functional impairment (Rankin Scale) [3] were employed as covariates in the regression analyses (Table 1). The robustness of the results was then tested again in a sensitivity analysis, with adjusting for a range of additional covariates that may contribute to the development of delirium. Medical history (diabetes, hypertension, coronary heart disease, hyperlipidemia, stroke, myocardial infarction, renal insufficiency), physical activity, physiological markers (blood pressure, total cholesterol, LDL, HDL, triglyceride, serum glucose, HbA1c, hsCRP, creatinine), BMI and ABI were controlled for. A Cox proportional hazards regression was performed to establish the risk of delirium for current smokers and quitters compared to lifelong abstainers after adjusting for other risk factors. In the first block, a trichotomous variable indicating the current smoking status at baseline (abstainer, quitter, current smoker), as well as age in years and sex, was simultaneously entered as predictors. In the second block, a forward selection method (Wald) was used to select the significant covariates of the variables shown in Table 1. As for all analyses, the dichotomous outcome variable indicated whether or not a participant received a diagnosis of delirium within the 6 years of the trial. For cases with incident delirium, the time variable was defined by the time between the date of the baseline examination and the date of the first diagnosis. For cases without incident delirium, the time variable Table 1 Established risk factors used as covariates in the Cox proportional hazards regression models and incident delirium according to smoking status at baseline. Descriptive
Smoking Abstainers N=2516
Age, mean (S.D.)⁎ Sex, N (%)⁎ Female Male Cognitive impairment (6CIT), N (%)⁎ 0–7 errors 8–12 errors 13–28 errors Depressive symptoms (GDS), N (%) b6 symptoms ≥6 symptoms Alcohol consumption (GP), N (%)⁎ No alcohol 1–14 drinks per week ≥15 drinks per week Use of psychoactive drugs (GP), N (%) No Yes Functional impairment (Rankin Scale), N (%) None (0) None despite symptoms (1) Slight to severe (2–5) Incident delirium, N (%) No Yes
Quitters N=865
Smokers N=373
68.3 (7.9)
67.5 (7.6)
63.9 (6.1)
1875 (74) 641 (26)
237 (27) 628 (73)
127 (34) 246 (66)
2265 (90) 196 (8) 55 (2)
762 (88) 95 (11) 8 (1)
326 (87) 35 (9) 12 (3)
2279 (91) 237 (9)
786 (91) 79 (9)
327 (88) 46 (12)
1146 (45) 1328 (53) 42 (2)
176 (20) 626 (73) 63 (7)
108 (29) 212 (57) 53 (14)
2210 (88) 306 (12)
782 (90) 83 (10)
327 (88) 46 (12)
2006 (80) 337 (13) 173 (7)
678 (88) 126 (15) 61 (7)
295 (79) 50 (13) 28 (8)
2476 (98) 40 (2)
854 (99) 11 (1)
363 (97) 10 (3)
Note: P values were calculated with analysis of variance for age and the χ2 test for the remaining variables. ⁎ Significant with Pb.05.
362
J.B. Hessler et al. / General Hospital Psychiatry 37 (2015) 360–364
was defined by the time between the date of the baseline examination and the date of study end, dropout or death. Using only the data of current smokers and quitters, a second Cox proportional hazards regression was performed to examine whether current smoking status or the amount smoked predict the incidence of delirium. In the first block, the risk factors chosen in the previous Cox proportional hazards regression, as well as age in years and sex, were forced into the model. In the second block, a forward inclusion method (Wald) was employed to select between smoking status (current smoker or quitter) and the number of pack-years (continuous) as better predictor of incident delirium. Results of the Cox proportional hazards regressions are reported as hazard ratios (HRs) and their 95% confidence intervals (95% CIs). Data analysis was performed with SPSS 20 for Macintosh.
3. Results At baseline, 3908 patients were examined. One-hundred fifty-four were excluded due to missing data, that is, 3754 [2239 (59.6%) female] were included. Three-hundred seventy-three (10.0%) were smokers at baseline, 865 (23.0%) were quitters and 2516 (67.0%) were lifelong abstainers. Mean pack-years of smokers and quitters were 23.8 (S.D.= 22.4). Over the course of the study, 61 participants (1.6%) received a diagnosis of delirium while being treated in the hospital. Descriptive statistics, risk factors and delirium incidence are listed according to smoking status at baseline in Table 1. Table 2 shows the results of the Cox proportional hazards regression model that was used to compare the risk of delirium for abstainers with the risk for quitters and current smokers after adjusting for several covariates. The forward selection method (Wald) included cognitive impairment, depressive symptoms and alcohol consumption as significant covariates next to age and sex, which were forced into the regression model. Functional impairment and use of psychotropic drugs [omnibus residual χ2(3)=3.78, P=.287] were not included in the model. In order to investigate whether current smoking or the lifetime amount smoked were associated with delirium, only data of current smokers or quitters were considered in a further Cox proportional hazards regression. After adjusting for the covariates selected in the previous model (Table 2), as well as age and sex, the forward selection method (Wald) included smoking status but not the number of packyears in the predictive model. Compared to quitters, current smokers had an increased risk of developing delirium (HR=3.22, 95% CI 1.20–8.62). The number of pack-years was not significantly associated with incident delirium and therefore not selected [residual χ 2(1)= 0.25, P= .874]. Reversing the order of smoking status and pack-years in the stepwise selection yielded the same results. Table 2 Risk of inpatient delirium for smokers compared to nonsmokers: fully adjusted Cox proportional hazards model. Risk factor Smoking at baseline examination Abstainers Quitters Smokers Age in years Male sex Cognitive impairment (6CIT) 0–7 errors 8–12 errors 13–28 errors ≥6 depressive symptoms (GDS) Alcohol consumption (GP) No alcohol 1–14 drinks per week ≥15 drinks per week a
HR
95% CI
1.0a 0.79 2.87 1.13 1.62
– 0.37–1.72 1.24–6.66 1.10–1.17 0.81–3.23
1.0* 1.36 2.96 2.13
– 0.67–2.76 1.27–6.92 1.14–3.98
1.0* 0.64 2.25
– 0.35–1.17 0.79–6.43
Reference categories are only shown for trichotomous variables.
The abovementioned associations remained unchanged when they were retested in an exploratory analysis with adjusting for a range of additional covariates, including cardiovascular risk factors and physiological markers. 4. Discussion Evidence on the association between smoking in older people and the risk of inpatient delirium is sparse. As the first to study delirium in the general population, we analyzed the data of almost 4000 older general practice patients who were prospectively followed for 6 years on average. The toxic agents in cigarette smoke induce atherosclerotic and microvascular changes, which build up gradually with time and remain even years after quitting [20]. Especially in older smokers, these changes are linked to cognitive decline and dementia [30,31], as well as vascular disease [32], which increases the risk of developing delirium [18]. Since these changes should be present in both smokers and quitters, this notion is challenged by the results of the present study. With regard to our research questions, we found that (1) smokers had a risk almost threefold higher than nonsmokers, whereas quitters and abstainers did not differ in their risk. (2) Current smoking but not the number of packyears was associated with the risk of developing delirium. In sum, it seems that only current smoking poses a risk of delirium, possibly due to the adverse consequences of nicotine withdrawal during hospital admission. Alcohol consumption was not associated with delirium, which might be due to the fact that diagnoses of “substance withdrawal with delirium” were not used as endpoints. Over the course of our study, 1.6% of the participants were diagnosed with delirium, contrasting other studies that reported higher rates in medical inpatients [5]. Several factors may have contributed to this difference. First, our study investigated the incidence of inpatient delirium in a sample from the general population, in which most individuals were not admitted to the hospital. Other studies investigated delirium only in hospitalized individuals, who are at high risk per se and, therefore, are expected to have higher incidence rates than in our study. Second, considering that the INVADE study is a prevention program, it is possible that rather health-conscious individuals decided to participate. The baseline risk of this group is presumably lower than in selected patient groups. Third, with a mean age around 68 years, our sample was young compared to other studies investigating delirium among older inpatients [4]. As age is an important risk factor for delirium, our sample would be expected to be at lower risk. Fourth, delirium is difficult to detect and often remains unrecognized in hospital patients [3]. As a consequence, the ICD codes we used as endpoints might not fully reflect all cases of inpatient delirium that occurred during our study. The design of our study allows only for speculating about the underlying mechanism leading from current smoking to delirium. Nicotine is thought to exert its neuroregulatory effects mainly by binding to nicotinic acetylcholine receptors (nAChR), which, in turn, affect other neurotransmitters such as dopamine, glutamate, γ-aminobutyric acid, serotonin and opioid peptides throughout the brain [33–36]. Chronic nicotine consumption can lead to nAChR desensitization and subsequent up-regulation (i.e., increase in the number of receptors) in order to maintain ACh homeostasis [36,37]. Contrary, nicotine withdrawal has been shown to cause ACh deficiency [33,37]. This state of ACh deficiency can lead to delirium [38,39], as has been demonstrated by the association between delirium and the use of anticholinergic drugs [40,41]. During acute nicotine withdrawal, ACh deficiency might be caused by desensitized and upregulated nAChRs that suddenly lack the sufficient amount of nicotinic stimulation in order to maintain ACh homeostasis. Evidence from neuroimaging studies suggests that this state of increased vulnerability might resolve within weeks when nicotine abstinence continues [37]. Consequently, quitters would not be expected to differ from abstainers in their risk of delirium after a relatively short time, as found in the present study. Importantly, this mechanism does not exclude other
J.B. Hessler et al. / General Hospital Psychiatry 37 (2015) 360–364
pathways from smoking to delirium (e.g., atherosclerosis, hypoxemia, increased carbon monoxide levels, smoke toxins). It is possible that multiple mechanisms work simultaneously in the pathogenesis of delirium in smokers. The supporting evidence in the literature for this mechanism of nicotine-withdrawal-induced ACh deficiency is only anecdotal. In a case report, a patient developed “delirium with psychosis” after discontinuing varenicline, a nAChR agonist and nicotine addiction treatment [42]. Three further case studies of hospital patients who had to interrupt heavy smoking habits due to their treatment reported quick remission of delirium after the application of nicotine patches [21–23]. The present study was the first to investigate the association between smoking and inpatient delirium in the older general population, which determines both its strengths and limitations. Delirium was diagnosed by the respective physicians in the hospital, who were blind to the results of the baseline examination, precluding any confirmation bias in the testing of our hypotheses. Smoking habits, as well as other risk factors, were assessed during routine examinations by GPs and diagnoses of inpatient delirium were retrieved from health insurance claims data. Our study is, therefore, well-embedded in the German health care system and possesses high validity within this framework. As mentioned before, the use of claims data as endpoints is not unproblematic, as some cases might remain undetected. Assuming, however, that a patient’s smoking habits might not be related to whether delirium is recognized by physicians, no bias would be expected to result from employing ICD codes as endpoints. Rather, this limitation should attenuate the association between smoking and inpatient delirium, which was nevertheless significant in our analyses. In addition, ICD codes assure some form of standardized and criterion-based diagnosis of delirium. Data related to the hospital admissions, such as ICU stay, surgery and anesthesia, were not collected and could not be controlled for in the statistical analyses. Owing to the extensive baseline examination, we were still able to adjust the analyses for a range of important risk factors. The incidence of delirium was predicted from baseline smoking habits. It is possible that some participants changed their habits over the course of the study. Also, it has to be mentioned that of the initially contacted insurants roughly two thirds did not decide to participate. Given that the INVADE trial is a prevention program for stroke and dementia, it seems plausible that rather health-oriented individuals enrolled. Whereas this fact may limit the degree to which our sample represents the general older population, it should not have affected the present study, which aimed to investigate a possible link between smoking and delirium in older people. However, together with the fact that there were only few smokers in the higher age groups, it might explain the relatively low number of smokers in our sample. In sum, our study’s limitations highlight the problems of examining a disease as complex, elusive and transitory as delirium on the population level. 5. Conclusion In due consideration of its limitations, the present study points to an independent link between smoking in the older population and inpatient delirium. The proposed mechanism of nicotine-withdrawalinduced ACh deficiency leading to delirium requires more investigation. For example, a randomized controlled trial could investigate the effectiveness of nicotine patches in reducing delirium in hospitalized smokers. If proven effective, nicotine patches could be easily integrated in existing prevention programs and could potentially improve patient outcomes as well as reduce costs. Funding Source This work was supported primarily by the health insurance company AOK Bayern. Further support for different time periods came from German Stroke Foundation, Bayer Vital GmbH, Berlin-Chemie AG,
363
Organon Pharmaceuticals, Ratiopharm GmbH, Sanofi-Synthelabo GmbH and TEVA Pharmaceutical Industries Ltd. Author Contributions HB, MB, TE, OG, HF, HP and DS designed, planned and conducted the study. JH and HB analyzed the data and wrote the manuscript. All authors provided important intellectual content during the writing process and approved the final version. Sponsor’s Role The content is solely the responsibility of the authors. The sponsors had no role in the design, method, subject recruitment, data collection, analysis and preparation of paper References [1] Inouye SK. Delirium in older persons. N Engl J Med 2006;354:1157–65. http://dx.doi. org/10.1056/NEJMra052321. [2] Inouye SK. Predisposing and precipitating factors for delirium in hospitalized older patients. Dement Geriatr Cogn Disord 1999;10:393–400. http://dx.doi.org/10.1159/ 000017177. [3] Inouye SK, Westendorp RG, Saczynski JS. Delirium in elderly people. Lancet 2014; 383:911–22. http://dx.doi.org/10.1016/S0140-6736(13)60688-1. [4] Ahmed S, Leurent B, Sampson EL. Risk factors for incident delirium among older people in acute hospital medical units: a systematic review and meta-analysis. Age Ageing 2014;43:326–33. http://dx.doi.org/10.1093/ageing/afu022. [5] Inouye SK. Delirium in hospitalized older patients. Clin Geriatr Med 1998;14: 745–64. [6] Witlox J, Eurelings LS, de Jonghe JF, Kalisvaart KJ, Eikelenboom P, et al. Delirium in elderly patients and the risk of postdischarge mortality, institutionalization, and dementia: a meta-analysis. JAMA 2010;304:443–51. http://dx.doi.org/10.1001/jama. 2010.1013. [7] Holroyd-Leduc JM, Khandwala F, Sink KM. How can delirium best be prevented and managed in older patients in hospital? CMAJ 2010;182:465–70. http://dx.doi.org/ 10.1503/cmaj.080519. [8] Milisen K, Lemiengre J, Braes T, Foreman MD. Multicomponent intervention strategies for managing delirium in hospitalized older people: systematic review. J Adv Nurs 2005;52:79–90. http://dx.doi.org/10.1111/j.1365-2648.2005.03557.x. [9] Hsieh SJ, Shum M, Lee AN, Hasselmark F, Gong MN. Cigarette smoking as a risk factor for delirium in hospitalized and intensive care unit patients. A systematic review. Ann Am Thorac Soc 2013;10:496–503. [10] Santos FS, Velasco IT, Frguas R. Risk factors for delirium in the elderly after coronary artery bypass graft surgery. Int Psychogeriatr 1999;16:175–93. http://dx.doi.org/10. 1017/S1041610204000365. [11] Van Rompaey B, Elseviers MM, Schuurmans MJ, Shortridge-Baggett LM, Truijen S, et al. Risk factors for delirium in intensive care patients: a prospective cohort study. Crit Care 2009;13:R77. http://dx.doi.org/10.1186/cc7892. [12] Sharma A, Malhotra S, Grover S, Jindal SK. Incidence, prevalence, risk factor and outcome of delirium in intensive care unit: a study from India. Gen Hosp Psychiatry 2012;34:639–46. http://dx.doi.org/10.1016/j.genhosppsych.2012.06.009. [13] Miyazaki S, Yoshitani K, Miura N, Irie T, Inatomi Y, et al. Risk factors of stroke and delirium after off-pump coronary artery bypass surgery. Interact Cardiovasc Thorac Surg 2011;12:379–83. http://dx.doi.org/10.1510/icvts.2010.248872. [14] Rudolph JL, Jones RN, Rasmussen LS, Silverstein JH, Inouye SK, et al. Independent vascular and cognitive risk factors for postoperative delirium. Am J Med 2007;120: 807–13. http://dx.doi.org/10.1016/j.amjmed.2007.02.026. [15] Large MC, Reichard C, Williams JTB, Chang C, Prasad S, et al. Incidence, risk factors, and complications of postoperative delirium in elderly patients undergoing radical cystectomy. Urology 2013;81:123–9. http://dx.doi.org/10.1016/j.urology. 2012.07.086. [16] Benoit AG, Campbell BI, Tanner JR, Staley JD, Wallbridge HR, et al. Risk factors and prevalence of perioperative cognitive dysfunction in abdominal aneurysm patients. J Vasc Surg 2005;42:884–90. http://dx.doi.org/10.1016/j.jvs.2005.07.032. [17] Huai J, Ye X. A meta-analysis of critically ill patients reveals several potential risk factors for delirium. Gen Hosp Psychiatry 2014;36:488–96. http://dx.doi.org/10.1016/j. genhosppsych.2014.05.002. [18] Pol RA, van Leeuwen BL, Reijnen MM, Zeebregts CJ. The relation between atherosclerosis and the occurrence of postoperative delirium in vascular surgery patients. Vasc Med 2012;17:116–22. http://dx.doi.org/10.1177/1358863X11429723. [19] Noimark D. Predicting the onset of delirium in the post-operative patient. Age Ageing 2009;38:368–73. http://dx.doi.org/10.1093/ageing/afp024. [20] Witteman JC, Grobbee DE, Valkenburg HA, van Hemert AM, Stijnen T, et al. Cigarette smoking and the development and progression of aortic atherosclerosis. A 9-year population-based follow-up study in women. Circulation 1993;88:2156–62. http:// dx.doi.org/10.1161/01.CIR.88.5.2156. [21] Gallagher RE. Nicotine withdrawal as an etiologic factor in delirium. J Pain Symptom Manage 1998;15:76–7. http://dx.doi.org/10.1016/S0885-3924(98)00047-5. [22] Krajnik M, Zylicz Z. Terminal restlessness and nicotine withdrawal. Lancet 1995;346: 1044. http://dx.doi.org/10.1016/S0140-6736(95)91729-2.
364
J.B. Hessler et al. / General Hospital Psychiatry 37 (2015) 360–364
[23] Mayer SA, Chong JY, Ridgway E, Min KC, Commichau C, et al. Delirium from nicotine withdrawal in neuro-ICU patients. Neurology 2001;57:551–3. http://dx.doi.org/10. 1212/WNL.57.3.551. [24] García Thuring L, Martínez Vigo M, Iruela Cuadrado L. Síndrome confusional tras interrupción brusca de consumo de nicotina. Med Clin (Barc) 2004;122:78–9. [25] Bickel H, Ander KH, Bronner M, Etgen T, Gnahn H, et al. Reduction of long-term care dependence after an 8-year primary care prevention program for stroke and dementia: the INVADE trial. J Am Heart Assoc 2012;1:1–11. http://dx.doi.org/10.1161/ JAHA.112.000786 [e000786]. [26] Yesavage JA, Sheikh JI. 9/Geriatric Depression Scale (GDS): recent evidence and development of a shorter violence. Clin Gerontol 1986;5:165–73. http://dx.doi.org/10. 1300/J018v05n01_09. [27] Bonita R, Beaglehole R. Recovery of motor function after stroke. Stroke 1988;19: 1497–500. http://dx.doi.org/10.1161/01.STR.19.12.1497. [28] Katzman R, Brown T, Fuld P, Peck A, Schechter R, et al. Validation of a short orientation-memory-concentration test of cognitive impairment. Am J Psychiatry 1983;140:734–9. [29] Brooke P, Bullock R. Validation of a 6 item cognitive impairment test with a view to primary care usage. Int J Geriatr Psychiatry 1999;14:936–40. http://dx.doi.org/10. 1002/(SICI)1099-1166(199911)14:11b936::AID-GPS39N3.0.CO;2-1. [30] Peters R, Poulter R, Warner J, Beckett N, Burch L, et al. Smoking, dementia and cognitive decline in the elderly, a systematic review. BMC Geriatr 2008;8:36. http://dx. doi.org/10.1186/1471-2318-8-36. [31] Anstey KJ, Sanden von C, Salim A, O’Kearney R. Smoking as a risk factor for dementia and cognitive decline: a meta-analysis of prospective studies. Am J Epidemiol 2007; 166:367–78. http://dx.doi.org/10.1093/aje/kwm116. [32] Ambrose JA, Barua RS. The pathophysiology of cigarette smoking and cardiovascular disease. J Am Coll Cardiol 2004;43:1731–7. http://dx.doi.org/10.1016/j.jacc.2003.12.047.
[33] Watkins SS, Koob GF, Markou A. Neural mechanisms underlying nicotine addiction: acute positive reinforcement and withdrawal. Nicotine Tob Res 2000;2:19–37. http://dx.doi.org/10.1080/14622200050011277. [34] Pomerleau OF, Pomerleau CS. Neuroregulators and the reinforcement of smoking: towards a biobehavioral explanation. Neurosci Biobehav Rev 1984;8:503–13. http://dx.doi.org/10.1016/0149-7634(84)90007-1. [35] Pomerleau OF, Rosecrans J. Neuroregulatory effects of nicotine. Psychoneuroendocrinology 1989;14:407–23. http://dx.doi.org/10.1016/0306-4530(89)90040-1. [36] De Biasi M, Dani JA. Reward, addiction, withdrawal to nicotine. Annu Rev Neurosci 2011;34:105–30. http://dx.doi.org/10.1146/annurev-neuro-061010-113734. [37] Jasinska AJ, Zorick T, Brody AL, Stein EA. Dual role of nicotine in addiction and cognition: a review of neuroimaging studies in humans. Neuropharmacology 2014; 84:111–22. http://dx.doi.org/10.1016/j.neuropharm.2013.02.015. [38] Hshieh TT, Fong TG, Marcantonio ER, Inouye SK. Cholinergic deficiency hypothesis in delirium: a synthesis of current evidence. J Gerontol A Biol Sci Med Sci 2008;63: 764–72. http://dx.doi.org/10.1513/AnnalsATS.201301-001OC. [39] Maldonado JR. Neuropathogenesis of delirium: review of current etiologic theories and common pathways. Am J Geriatr Psychiatry 2013;21:1190–222. http://dx.doi. org/10.1016/j.jagp.2013.09.005. [40] Praticò C, Quattrone D, Lucanto T, Amato A, Penna O, et al. Drugs of anesthesia acting on central cholinergic system may cause post-operative cognitive dysfunction and delirium. Med Hypotheses 2005;65:972–82. http://dx.doi.org/10.1016/j.mehy. 2005.05.037. [41] Karlsson I. Drugs that induce delirium. Dement Geriatr Cogn Disord 1999;10:412–5. http://dx.doi.org/10.1159/000017180. [42] May AC, Rose D. Varenicline withdrawal-induced delirium with psychosis. Am J Psychiatry 2010;167:720–1. http://dx.doi.org/10.1176/appi.ajp.2010. 10010020.
Contents lists available at ScienceDirect
General Hospital Psychiatry journal homepage: http://www.ghpjournal.com
Smoking increases the risk of delirium for older inpatients: a prospective population-based study Johannes Baltasar Hessler, M.Sc. a,⁎, Monika Brönner, M.D. a, Thorleif Etgen, M.D. a,b, Othmar Gotzler, M.D. c, Hans Förstl, M.D. a, Holger Poppert, M.D. d, Dirk Sander, M.D. e, Horst Bickel, Ph.D. a a
Department of Psychiatry and Psychotherapy, Technische Universität München Klinikum Rechts der Isar, Ismaninger Strasse 22, 81675 Munich, Germany Department of Neurology, Kliniken Südostbayern–Klinikum Traunstein, Cuno-Niggl-Strasse 3, 83278 Traunstein, Germany INVADE Study Group, Karl-Böhm-Strasse 32, 85598 Baldham, Germany d Department of Neurology, Technische Universität München Klinikum Rechts der Isar, Ismaninger Strasse 22, 81675 Munich, Germany e Department of Neurology, Benedictus Krankenhaus Tutzing, Bahnhofstrasse 5, 82327 Tutzing, Germany b c
a r t i c l e
i n f o
Article history: Received 26 November 2014 Revised 13 March 2015 Accepted 13 March 2015 Keywords: Acetylcholine Nicotine Withdrawal Prevention Hospital
a b s t r a c t Objectives: To investigate the association between smoking in the older population and the risk of inpatient delirium, which is common and has adverse consequences. Method: Participants (N=3754) were insurants aged ≥55 years of the largest German statutory health insurance company, who enrolled in a 6-year prospective population-based study. Baseline smoking, adjusted for age, sex, depressive symptoms, cognitive impairment and alcohol consumption, was analyzed as risk factor of inpatient delirium. Results are presented as hazard ratios (HRs) and 95% confidence intervals (95% CIs). Results: Three-hundred seventy-three (10.0%) participants were smokers at baseline, 865 (23.0%) were quitters and 2516 (67.0%) were lifelong abstainers. Mean pack-years of smokers and quitters were 23.8 (S.D.=22.4). Sixty-one (1.6%) received a diagnosis of inpatient delirium. Smokers had an increased risk of delirium compared to abstainers in the fully adjusted model (HR=2.87, 95% CI 1.24–6.66). Quitters and abstainers did not differ (HR=0.79, 95% CI 0.37–1.72). Comparing smokers and quitters, current smoking status (HR=3.22, 95% CI 1.20–8.62) but not pack-years [residual χ2(1)=0.25, P=.874] were associated with inpatient delirium. Conclusion: Only current smoking but not being a quitter and the lifetime amount smoked were associated with inpatient delirium, indicating that acute nicotine withdrawal may represent a relevant pathogenic mechanism. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Delirium describes acute brain failure that causes impairment of attention and other cognitive functions [1]. The occurrence of delirium appears to result from the interplay of a wide range of precipitating noxious events and predisposing factors [2]. Among others, high age; cognitive, functional and sensory impairment; infection; physiological abnormalities; a history of physical and mental disease; drug use; alcohol misuse; physical constraints; anesthesia; use of a bladder catheter; urgent or trauma admission and intensive care unit (ICU) stay can contribute to the etiology of delirium [3,4]. Delirium is most common among older hospital patients [5], with the highest incidence rates in surgical and ICU patients [3]. Given the high costs, prolonged hospital stays and increased rates of morbidity and mortality associated with ⁎ Corresponding author. Department of Psychiatry and Psychotherapy, Klinikum Rechts der Isar der Technische Universität München, Ismaninger Strasse 22, 81675 Munich, Germany. Tel.: +49-89-4140-6183; fax: +49-89-4140-6379. E-mail addresses: [email protected] (J.B. Hessler), [email protected] (M. Brönner), [email protected] (T. Etgen), [email protected] (O. Gotzler), [email protected] (H. Förstl), [email protected] (H. Poppert), [email protected] (D. Sander), [email protected] (H. Bickel). http://dx.doi.org/10.1016/j.genhosppsych.2015.03.009 0163-8343/© 2015 Elsevier Inc. All rights reserved.
delirium [1], as well as the poor cognitive and functional prognosis for those affected [6], the investigation of means to reduce its incidence in older hospital patients is of vital importance. Several multicomponent prevention strategies have been proven effective in reducing the incidence of delirium in hospitalized older people [7,8]. Surprisingly, since being a common habit, smoking has not been considered in this context. This neglect is possibly due to the sparse and inconsistent evidence on the role of smoking in the pathogenesis of delirium [9]. In some studies, current smoking was found to increase the risk of delirium in the hospital [10–13], yet others found no effect [14–17]. Despite the lack of clear evidence, there is a reason to believe that smoking could be a relevant factor in the development of delirium, both as predisposing factor and precipitating noxious event. Next to other factors (e.g., hypoxemia, increased carbon monoxide levels), smoking potentially increases the risk of delirium through the buildup of microvascular and atherosclerotic changes in the brain [11,18,19]. These changes are positively correlated with the number of cigarettes smoked and can persist for many years after quitting [20]. If vascular changes were the main agent behind the association between smoking and delirium, not current smoking status but the lifetime number of cigarettes smoked should predict the occurrence of delirium. Case
J.B. Hessler et al. / General Hospital Psychiatry 37 (2015) 360–364
reports of heavily smoking hospital patients [21–24], however, point to an alternative or additional pathway from smoking to delirium. These patients had to stop smoking due to their treatment and developed delirium. The symptoms remitted within hours after the application of transdermal nicotine replacement. These reports suggest that acute nicotine withdrawal due to smoking cessation in the hospital may constitute a noxious event, which directly relates to the development of delirium. If nicotine withdrawal and not atherosclerosis were the critical factor linking smoking and delirium, current smoking status but not the number of cigarettes smoked should predict incident delirium. The present study aimed to investigate the unclear association between smoking in the older population and the incidence of inpatient delirium, as well as explore possible underlying mechanisms. To this purpose, we analyzed data of almost 4000 individuals older than 54 years who were prospectively observed for an average of 6 years over the course of the INVADE trial (Intervention Project on Cerebrovascular Disease and Dementia in the District of Ebersberg [25]). In particular, we examined two questions. (1) Do quitters and current smokers differ from abstainers in their risk of inpatient delirium? (2) Is smoking status or the lifetime amount smoked the better predictor of inpatient delirium among current smokers and quitters? 2. Method 2.1. Participants The INVADE trial is a prospective and population-based cohort study in a geographically defined area. Participants were identified from the database of the statutory health insurance company Allgemeine Ortskrankenkasse (AOK). Membership in a health insurance is mandatory in Germany and AOK holds the largest market share, representing around 40% of the total population. In 2001, 11,317 insurants met the inclusion criteria of being 55 years or older, as well as living in the district of Ebersberg, and were invited to participate. Between 2001 and 2003, 3908 participants enrolled in the trial. 2.2. Procedure The ethics committee of the Faculty of Medicine at the Technische Universität München approved the study protocol and all participants signed informed consent. The participants were examined by their respective general practitioner (GP) at baseline and at follow-ups after 2, 4 and 6 years. At baseline, GPs asked the participants about their current and previous smoking habits. Based on this information, the participants were categorized into current smokers, quitters and lifelong abstainers. For smokers and quitters, the number of pack-years was determined as an indicator of the lifetime amount smoked. Pack-years are calculated by multiplying the number of packs a person smoked per day with the number of years the person has smoked. Further, the participants filled in questionnaires about sociodemographic data, subjective health, use of medical services, memory complaints and depressive symptoms (Geriatric Depression Scale or GDS [26]). GPs reported the patients’ previous and current diagnoses, current medication, alcohol consumption, physical activity, body mass index (BMI), impairment of activities of daily living (Rankin Scale [27]), blood pressure, ankle-to-brachial index (ABI), and cognitive status (6-Item Cognitive Impairment Test or 6CIT [28,29]) and conducted an electrocardiogram. GPs also took fasting blood samples that were analyzed in a central laboratory with regard to total cholesterol, low- and high-density lipoprotein (LDL, HDL) cholesterol, triglycerides, serum glucose, glycosylated hemoglobin A1c (HbA1c), creatinine, homocysteine and high-sensitivity C-reactive protein (hsCRP). A detailed description of the INVADE trial and the baseline examination is published [25]. Cases of delirium not induced by alcohol and other psychoactive substances were identified by searching claims data of the AOK health
361
insurance for International Statistical Classification of Diseases (ICD), 10th Revision codes F05.0–F05.9 that were assigned during a hospital stay over the course of the 6-year period of the INVADE trial. If participants received more than one diagnosis of delirium over the course of the study, the first diagnosis was used as endpoint. Diagnoses of substance withdrawal state with delirium (F1×.4) were not considered.
2.3. Statistical analysis In order to establish an independent association between smoking and inpatient delirium, the known risk factors age, sex, cognitive impairment (6CIT), depressive symptoms (GDS), alcohol consumption (GP), use of psychoactive drugs (GP) and functional impairment (Rankin Scale) [3] were employed as covariates in the regression analyses (Table 1). The robustness of the results was then tested again in a sensitivity analysis, with adjusting for a range of additional covariates that may contribute to the development of delirium. Medical history (diabetes, hypertension, coronary heart disease, hyperlipidemia, stroke, myocardial infarction, renal insufficiency), physical activity, physiological markers (blood pressure, total cholesterol, LDL, HDL, triglyceride, serum glucose, HbA1c, hsCRP, creatinine), BMI and ABI were controlled for. A Cox proportional hazards regression was performed to establish the risk of delirium for current smokers and quitters compared to lifelong abstainers after adjusting for other risk factors. In the first block, a trichotomous variable indicating the current smoking status at baseline (abstainer, quitter, current smoker), as well as age in years and sex, was simultaneously entered as predictors. In the second block, a forward selection method (Wald) was used to select the significant covariates of the variables shown in Table 1. As for all analyses, the dichotomous outcome variable indicated whether or not a participant received a diagnosis of delirium within the 6 years of the trial. For cases with incident delirium, the time variable was defined by the time between the date of the baseline examination and the date of the first diagnosis. For cases without incident delirium, the time variable Table 1 Established risk factors used as covariates in the Cox proportional hazards regression models and incident delirium according to smoking status at baseline. Descriptive
Smoking Abstainers N=2516
Age, mean (S.D.)⁎ Sex, N (%)⁎ Female Male Cognitive impairment (6CIT), N (%)⁎ 0–7 errors 8–12 errors 13–28 errors Depressive symptoms (GDS), N (%) b6 symptoms ≥6 symptoms Alcohol consumption (GP), N (%)⁎ No alcohol 1–14 drinks per week ≥15 drinks per week Use of psychoactive drugs (GP), N (%) No Yes Functional impairment (Rankin Scale), N (%) None (0) None despite symptoms (1) Slight to severe (2–5) Incident delirium, N (%) No Yes
Quitters N=865
Smokers N=373
68.3 (7.9)
67.5 (7.6)
63.9 (6.1)
1875 (74) 641 (26)
237 (27) 628 (73)
127 (34) 246 (66)
2265 (90) 196 (8) 55 (2)
762 (88) 95 (11) 8 (1)
326 (87) 35 (9) 12 (3)
2279 (91) 237 (9)
786 (91) 79 (9)
327 (88) 46 (12)
1146 (45) 1328 (53) 42 (2)
176 (20) 626 (73) 63 (7)
108 (29) 212 (57) 53 (14)
2210 (88) 306 (12)
782 (90) 83 (10)
327 (88) 46 (12)
2006 (80) 337 (13) 173 (7)
678 (88) 126 (15) 61 (7)
295 (79) 50 (13) 28 (8)
2476 (98) 40 (2)
854 (99) 11 (1)
363 (97) 10 (3)
Note: P values were calculated with analysis of variance for age and the χ2 test for the remaining variables. ⁎ Significant with Pb.05.
362
J.B. Hessler et al. / General Hospital Psychiatry 37 (2015) 360–364
was defined by the time between the date of the baseline examination and the date of study end, dropout or death. Using only the data of current smokers and quitters, a second Cox proportional hazards regression was performed to examine whether current smoking status or the amount smoked predict the incidence of delirium. In the first block, the risk factors chosen in the previous Cox proportional hazards regression, as well as age in years and sex, were forced into the model. In the second block, a forward inclusion method (Wald) was employed to select between smoking status (current smoker or quitter) and the number of pack-years (continuous) as better predictor of incident delirium. Results of the Cox proportional hazards regressions are reported as hazard ratios (HRs) and their 95% confidence intervals (95% CIs). Data analysis was performed with SPSS 20 for Macintosh.
3. Results At baseline, 3908 patients were examined. One-hundred fifty-four were excluded due to missing data, that is, 3754 [2239 (59.6%) female] were included. Three-hundred seventy-three (10.0%) were smokers at baseline, 865 (23.0%) were quitters and 2516 (67.0%) were lifelong abstainers. Mean pack-years of smokers and quitters were 23.8 (S.D.= 22.4). Over the course of the study, 61 participants (1.6%) received a diagnosis of delirium while being treated in the hospital. Descriptive statistics, risk factors and delirium incidence are listed according to smoking status at baseline in Table 1. Table 2 shows the results of the Cox proportional hazards regression model that was used to compare the risk of delirium for abstainers with the risk for quitters and current smokers after adjusting for several covariates. The forward selection method (Wald) included cognitive impairment, depressive symptoms and alcohol consumption as significant covariates next to age and sex, which were forced into the regression model. Functional impairment and use of psychotropic drugs [omnibus residual χ2(3)=3.78, P=.287] were not included in the model. In order to investigate whether current smoking or the lifetime amount smoked were associated with delirium, only data of current smokers or quitters were considered in a further Cox proportional hazards regression. After adjusting for the covariates selected in the previous model (Table 2), as well as age and sex, the forward selection method (Wald) included smoking status but not the number of packyears in the predictive model. Compared to quitters, current smokers had an increased risk of developing delirium (HR=3.22, 95% CI 1.20–8.62). The number of pack-years was not significantly associated with incident delirium and therefore not selected [residual χ 2(1)= 0.25, P= .874]. Reversing the order of smoking status and pack-years in the stepwise selection yielded the same results. Table 2 Risk of inpatient delirium for smokers compared to nonsmokers: fully adjusted Cox proportional hazards model. Risk factor Smoking at baseline examination Abstainers Quitters Smokers Age in years Male sex Cognitive impairment (6CIT) 0–7 errors 8–12 errors 13–28 errors ≥6 depressive symptoms (GDS) Alcohol consumption (GP) No alcohol 1–14 drinks per week ≥15 drinks per week a
HR
95% CI
1.0a 0.79 2.87 1.13 1.62
– 0.37–1.72 1.24–6.66 1.10–1.17 0.81–3.23
1.0* 1.36 2.96 2.13
– 0.67–2.76 1.27–6.92 1.14–3.98
1.0* 0.64 2.25
– 0.35–1.17 0.79–6.43
Reference categories are only shown for trichotomous variables.
The abovementioned associations remained unchanged when they were retested in an exploratory analysis with adjusting for a range of additional covariates, including cardiovascular risk factors and physiological markers. 4. Discussion Evidence on the association between smoking in older people and the risk of inpatient delirium is sparse. As the first to study delirium in the general population, we analyzed the data of almost 4000 older general practice patients who were prospectively followed for 6 years on average. The toxic agents in cigarette smoke induce atherosclerotic and microvascular changes, which build up gradually with time and remain even years after quitting [20]. Especially in older smokers, these changes are linked to cognitive decline and dementia [30,31], as well as vascular disease [32], which increases the risk of developing delirium [18]. Since these changes should be present in both smokers and quitters, this notion is challenged by the results of the present study. With regard to our research questions, we found that (1) smokers had a risk almost threefold higher than nonsmokers, whereas quitters and abstainers did not differ in their risk. (2) Current smoking but not the number of packyears was associated with the risk of developing delirium. In sum, it seems that only current smoking poses a risk of delirium, possibly due to the adverse consequences of nicotine withdrawal during hospital admission. Alcohol consumption was not associated with delirium, which might be due to the fact that diagnoses of “substance withdrawal with delirium” were not used as endpoints. Over the course of our study, 1.6% of the participants were diagnosed with delirium, contrasting other studies that reported higher rates in medical inpatients [5]. Several factors may have contributed to this difference. First, our study investigated the incidence of inpatient delirium in a sample from the general population, in which most individuals were not admitted to the hospital. Other studies investigated delirium only in hospitalized individuals, who are at high risk per se and, therefore, are expected to have higher incidence rates than in our study. Second, considering that the INVADE study is a prevention program, it is possible that rather health-conscious individuals decided to participate. The baseline risk of this group is presumably lower than in selected patient groups. Third, with a mean age around 68 years, our sample was young compared to other studies investigating delirium among older inpatients [4]. As age is an important risk factor for delirium, our sample would be expected to be at lower risk. Fourth, delirium is difficult to detect and often remains unrecognized in hospital patients [3]. As a consequence, the ICD codes we used as endpoints might not fully reflect all cases of inpatient delirium that occurred during our study. The design of our study allows only for speculating about the underlying mechanism leading from current smoking to delirium. Nicotine is thought to exert its neuroregulatory effects mainly by binding to nicotinic acetylcholine receptors (nAChR), which, in turn, affect other neurotransmitters such as dopamine, glutamate, γ-aminobutyric acid, serotonin and opioid peptides throughout the brain [33–36]. Chronic nicotine consumption can lead to nAChR desensitization and subsequent up-regulation (i.e., increase in the number of receptors) in order to maintain ACh homeostasis [36,37]. Contrary, nicotine withdrawal has been shown to cause ACh deficiency [33,37]. This state of ACh deficiency can lead to delirium [38,39], as has been demonstrated by the association between delirium and the use of anticholinergic drugs [40,41]. During acute nicotine withdrawal, ACh deficiency might be caused by desensitized and upregulated nAChRs that suddenly lack the sufficient amount of nicotinic stimulation in order to maintain ACh homeostasis. Evidence from neuroimaging studies suggests that this state of increased vulnerability might resolve within weeks when nicotine abstinence continues [37]. Consequently, quitters would not be expected to differ from abstainers in their risk of delirium after a relatively short time, as found in the present study. Importantly, this mechanism does not exclude other
J.B. Hessler et al. / General Hospital Psychiatry 37 (2015) 360–364
pathways from smoking to delirium (e.g., atherosclerosis, hypoxemia, increased carbon monoxide levels, smoke toxins). It is possible that multiple mechanisms work simultaneously in the pathogenesis of delirium in smokers. The supporting evidence in the literature for this mechanism of nicotine-withdrawal-induced ACh deficiency is only anecdotal. In a case report, a patient developed “delirium with psychosis” after discontinuing varenicline, a nAChR agonist and nicotine addiction treatment [42]. Three further case studies of hospital patients who had to interrupt heavy smoking habits due to their treatment reported quick remission of delirium after the application of nicotine patches [21–23]. The present study was the first to investigate the association between smoking and inpatient delirium in the older general population, which determines both its strengths and limitations. Delirium was diagnosed by the respective physicians in the hospital, who were blind to the results of the baseline examination, precluding any confirmation bias in the testing of our hypotheses. Smoking habits, as well as other risk factors, were assessed during routine examinations by GPs and diagnoses of inpatient delirium were retrieved from health insurance claims data. Our study is, therefore, well-embedded in the German health care system and possesses high validity within this framework. As mentioned before, the use of claims data as endpoints is not unproblematic, as some cases might remain undetected. Assuming, however, that a patient’s smoking habits might not be related to whether delirium is recognized by physicians, no bias would be expected to result from employing ICD codes as endpoints. Rather, this limitation should attenuate the association between smoking and inpatient delirium, which was nevertheless significant in our analyses. In addition, ICD codes assure some form of standardized and criterion-based diagnosis of delirium. Data related to the hospital admissions, such as ICU stay, surgery and anesthesia, were not collected and could not be controlled for in the statistical analyses. Owing to the extensive baseline examination, we were still able to adjust the analyses for a range of important risk factors. The incidence of delirium was predicted from baseline smoking habits. It is possible that some participants changed their habits over the course of the study. Also, it has to be mentioned that of the initially contacted insurants roughly two thirds did not decide to participate. Given that the INVADE trial is a prevention program for stroke and dementia, it seems plausible that rather health-oriented individuals enrolled. Whereas this fact may limit the degree to which our sample represents the general older population, it should not have affected the present study, which aimed to investigate a possible link between smoking and delirium in older people. However, together with the fact that there were only few smokers in the higher age groups, it might explain the relatively low number of smokers in our sample. In sum, our study’s limitations highlight the problems of examining a disease as complex, elusive and transitory as delirium on the population level. 5. Conclusion In due consideration of its limitations, the present study points to an independent link between smoking in the older population and inpatient delirium. The proposed mechanism of nicotine-withdrawalinduced ACh deficiency leading to delirium requires more investigation. For example, a randomized controlled trial could investigate the effectiveness of nicotine patches in reducing delirium in hospitalized smokers. If proven effective, nicotine patches could be easily integrated in existing prevention programs and could potentially improve patient outcomes as well as reduce costs. Funding Source This work was supported primarily by the health insurance company AOK Bayern. Further support for different time periods came from German Stroke Foundation, Bayer Vital GmbH, Berlin-Chemie AG,
363
Organon Pharmaceuticals, Ratiopharm GmbH, Sanofi-Synthelabo GmbH and TEVA Pharmaceutical Industries Ltd. Author Contributions HB, MB, TE, OG, HF, HP and DS designed, planned and conducted the study. JH and HB analyzed the data and wrote the manuscript. All authors provided important intellectual content during the writing process and approved the final version. Sponsor’s Role The content is solely the responsibility of the authors. The sponsors had no role in the design, method, subject recruitment, data collection, analysis and preparation of paper References [1] Inouye SK. Delirium in older persons. N Engl J Med 2006;354:1157–65. http://dx.doi. org/10.1056/NEJMra052321. [2] Inouye SK. Predisposing and precipitating factors for delirium in hospitalized older patients. Dement Geriatr Cogn Disord 1999;10:393–400. http://dx.doi.org/10.1159/ 000017177. [3] Inouye SK, Westendorp RG, Saczynski JS. Delirium in elderly people. Lancet 2014; 383:911–22. http://dx.doi.org/10.1016/S0140-6736(13)60688-1. [4] Ahmed S, Leurent B, Sampson EL. Risk factors for incident delirium among older people in acute hospital medical units: a systematic review and meta-analysis. Age Ageing 2014;43:326–33. http://dx.doi.org/10.1093/ageing/afu022. [5] Inouye SK. Delirium in hospitalized older patients. Clin Geriatr Med 1998;14: 745–64. [6] Witlox J, Eurelings LS, de Jonghe JF, Kalisvaart KJ, Eikelenboom P, et al. Delirium in elderly patients and the risk of postdischarge mortality, institutionalization, and dementia: a meta-analysis. JAMA 2010;304:443–51. http://dx.doi.org/10.1001/jama. 2010.1013. [7] Holroyd-Leduc JM, Khandwala F, Sink KM. How can delirium best be prevented and managed in older patients in hospital? CMAJ 2010;182:465–70. http://dx.doi.org/ 10.1503/cmaj.080519. [8] Milisen K, Lemiengre J, Braes T, Foreman MD. Multicomponent intervention strategies for managing delirium in hospitalized older people: systematic review. J Adv Nurs 2005;52:79–90. http://dx.doi.org/10.1111/j.1365-2648.2005.03557.x. [9] Hsieh SJ, Shum M, Lee AN, Hasselmark F, Gong MN. Cigarette smoking as a risk factor for delirium in hospitalized and intensive care unit patients. A systematic review. Ann Am Thorac Soc 2013;10:496–503. [10] Santos FS, Velasco IT, Frguas R. Risk factors for delirium in the elderly after coronary artery bypass graft surgery. Int Psychogeriatr 1999;16:175–93. http://dx.doi.org/10. 1017/S1041610204000365. [11] Van Rompaey B, Elseviers MM, Schuurmans MJ, Shortridge-Baggett LM, Truijen S, et al. Risk factors for delirium in intensive care patients: a prospective cohort study. Crit Care 2009;13:R77. http://dx.doi.org/10.1186/cc7892. [12] Sharma A, Malhotra S, Grover S, Jindal SK. Incidence, prevalence, risk factor and outcome of delirium in intensive care unit: a study from India. Gen Hosp Psychiatry 2012;34:639–46. http://dx.doi.org/10.1016/j.genhosppsych.2012.06.009. [13] Miyazaki S, Yoshitani K, Miura N, Irie T, Inatomi Y, et al. Risk factors of stroke and delirium after off-pump coronary artery bypass surgery. Interact Cardiovasc Thorac Surg 2011;12:379–83. http://dx.doi.org/10.1510/icvts.2010.248872. [14] Rudolph JL, Jones RN, Rasmussen LS, Silverstein JH, Inouye SK, et al. Independent vascular and cognitive risk factors for postoperative delirium. Am J Med 2007;120: 807–13. http://dx.doi.org/10.1016/j.amjmed.2007.02.026. [15] Large MC, Reichard C, Williams JTB, Chang C, Prasad S, et al. Incidence, risk factors, and complications of postoperative delirium in elderly patients undergoing radical cystectomy. Urology 2013;81:123–9. http://dx.doi.org/10.1016/j.urology. 2012.07.086. [16] Benoit AG, Campbell BI, Tanner JR, Staley JD, Wallbridge HR, et al. Risk factors and prevalence of perioperative cognitive dysfunction in abdominal aneurysm patients. J Vasc Surg 2005;42:884–90. http://dx.doi.org/10.1016/j.jvs.2005.07.032. [17] Huai J, Ye X. A meta-analysis of critically ill patients reveals several potential risk factors for delirium. Gen Hosp Psychiatry 2014;36:488–96. http://dx.doi.org/10.1016/j. genhosppsych.2014.05.002. [18] Pol RA, van Leeuwen BL, Reijnen MM, Zeebregts CJ. The relation between atherosclerosis and the occurrence of postoperative delirium in vascular surgery patients. Vasc Med 2012;17:116–22. http://dx.doi.org/10.1177/1358863X11429723. [19] Noimark D. Predicting the onset of delirium in the post-operative patient. Age Ageing 2009;38:368–73. http://dx.doi.org/10.1093/ageing/afp024. [20] Witteman JC, Grobbee DE, Valkenburg HA, van Hemert AM, Stijnen T, et al. Cigarette smoking and the development and progression of aortic atherosclerosis. A 9-year population-based follow-up study in women. Circulation 1993;88:2156–62. http:// dx.doi.org/10.1161/01.CIR.88.5.2156. [21] Gallagher RE. Nicotine withdrawal as an etiologic factor in delirium. J Pain Symptom Manage 1998;15:76–7. http://dx.doi.org/10.1016/S0885-3924(98)00047-5. [22] Krajnik M, Zylicz Z. Terminal restlessness and nicotine withdrawal. Lancet 1995;346: 1044. http://dx.doi.org/10.1016/S0140-6736(95)91729-2.
364
J.B. Hessler et al. / General Hospital Psychiatry 37 (2015) 360–364
[23] Mayer SA, Chong JY, Ridgway E, Min KC, Commichau C, et al. Delirium from nicotine withdrawal in neuro-ICU patients. Neurology 2001;57:551–3. http://dx.doi.org/10. 1212/WNL.57.3.551. [24] García Thuring L, Martínez Vigo M, Iruela Cuadrado L. Síndrome confusional tras interrupción brusca de consumo de nicotina. Med Clin (Barc) 2004;122:78–9. [25] Bickel H, Ander KH, Bronner M, Etgen T, Gnahn H, et al. Reduction of long-term care dependence after an 8-year primary care prevention program for stroke and dementia: the INVADE trial. J Am Heart Assoc 2012;1:1–11. http://dx.doi.org/10.1161/ JAHA.112.000786 [e000786]. [26] Yesavage JA, Sheikh JI. 9/Geriatric Depression Scale (GDS): recent evidence and development of a shorter violence. Clin Gerontol 1986;5:165–73. http://dx.doi.org/10. 1300/J018v05n01_09. [27] Bonita R, Beaglehole R. Recovery of motor function after stroke. Stroke 1988;19: 1497–500. http://dx.doi.org/10.1161/01.STR.19.12.1497. [28] Katzman R, Brown T, Fuld P, Peck A, Schechter R, et al. Validation of a short orientation-memory-concentration test of cognitive impairment. Am J Psychiatry 1983;140:734–9. [29] Brooke P, Bullock R. Validation of a 6 item cognitive impairment test with a view to primary care usage. Int J Geriatr Psychiatry 1999;14:936–40. http://dx.doi.org/10. 1002/(SICI)1099-1166(199911)14:11b936::AID-GPS39N3.0.CO;2-1. [30] Peters R, Poulter R, Warner J, Beckett N, Burch L, et al. Smoking, dementia and cognitive decline in the elderly, a systematic review. BMC Geriatr 2008;8:36. http://dx. doi.org/10.1186/1471-2318-8-36. [31] Anstey KJ, Sanden von C, Salim A, O’Kearney R. Smoking as a risk factor for dementia and cognitive decline: a meta-analysis of prospective studies. Am J Epidemiol 2007; 166:367–78. http://dx.doi.org/10.1093/aje/kwm116. [32] Ambrose JA, Barua RS. The pathophysiology of cigarette smoking and cardiovascular disease. J Am Coll Cardiol 2004;43:1731–7. http://dx.doi.org/10.1016/j.jacc.2003.12.047.
[33] Watkins SS, Koob GF, Markou A. Neural mechanisms underlying nicotine addiction: acute positive reinforcement and withdrawal. Nicotine Tob Res 2000;2:19–37. http://dx.doi.org/10.1080/14622200050011277. [34] Pomerleau OF, Pomerleau CS. Neuroregulators and the reinforcement of smoking: towards a biobehavioral explanation. Neurosci Biobehav Rev 1984;8:503–13. http://dx.doi.org/10.1016/0149-7634(84)90007-1. [35] Pomerleau OF, Rosecrans J. Neuroregulatory effects of nicotine. Psychoneuroendocrinology 1989;14:407–23. http://dx.doi.org/10.1016/0306-4530(89)90040-1. [36] De Biasi M, Dani JA. Reward, addiction, withdrawal to nicotine. Annu Rev Neurosci 2011;34:105–30. http://dx.doi.org/10.1146/annurev-neuro-061010-113734. [37] Jasinska AJ, Zorick T, Brody AL, Stein EA. Dual role of nicotine in addiction and cognition: a review of neuroimaging studies in humans. Neuropharmacology 2014; 84:111–22. http://dx.doi.org/10.1016/j.neuropharm.2013.02.015. [38] Hshieh TT, Fong TG, Marcantonio ER, Inouye SK. Cholinergic deficiency hypothesis in delirium: a synthesis of current evidence. J Gerontol A Biol Sci Med Sci 2008;63: 764–72. http://dx.doi.org/10.1513/AnnalsATS.201301-001OC. [39] Maldonado JR. Neuropathogenesis of delirium: review of current etiologic theories and common pathways. Am J Geriatr Psychiatry 2013;21:1190–222. http://dx.doi. org/10.1016/j.jagp.2013.09.005. [40] Praticò C, Quattrone D, Lucanto T, Amato A, Penna O, et al. Drugs of anesthesia acting on central cholinergic system may cause post-operative cognitive dysfunction and delirium. Med Hypotheses 2005;65:972–82. http://dx.doi.org/10.1016/j.mehy. 2005.05.037. [41] Karlsson I. Drugs that induce delirium. Dement Geriatr Cogn Disord 1999;10:412–5. http://dx.doi.org/10.1159/000017180. [42] May AC, Rose D. Varenicline withdrawal-induced delirium with psychosis. Am J Psychiatry 2010;167:720–1. http://dx.doi.org/10.1176/appi.ajp.2010. 10010020.